Ability to enable more detailled logging

### Description

Hi @joepavitt,

I was wondering last night why my injected messages did not arrive on my heatmap node. After going to the drawing table, my second version of my [cheat sheet](https://github.com/FlowFuse/node-red-dashboard/issues/404#issuecomment-1847899338) made things clear for me. So after a quick fix, I now finally receive messages.

But I have to admit that it is kind of difficult at the moment to find out what is going wrong. For example you cannot easily find a socketio related source file in the browser debugger, because all the libraries are now packaged inside some obscure index-xxxx.js files by Vite. Finally I managed to find and add a conditional breakpoint for my node:


This way at least I knew that my injected message arrived in my browser.

It would be nice if some day somehow some extra logging could be activated, which are disabled by default to avoid polluting the browser console log. So that you know a message arrives, or that the default onInput of your component will be called, or that there is nothing specified in the data tracker specified for you node (so your message will get lost), and so on... That way you can quickly see in the Developer tools of the browser what you are doing wrong.

Not sure what would be the best way to activate this kind of logging behaviour. Perhaps by injecting a special kind of message into a node, to switch it to trace level? And that you can specify the scope in that message (single node/all nodes)? Or perhaps there is already something related in place in the current dashboard? Don't know...

This is ***NOT*** urgent, but it would certainly help us in the future to develop UI nodes.

Thanks!!
Bart

### Have you provided an initial effort estimate for this issue?

I have provided an initial effort estimate

未指定文件或测试。首先跟踪浏览器端的 Socket.IO 消息流、data tracker 和默认的 onInput 路径，然后检查现有的 logging 控制。定义如何启用和限定 logging 的范围，并验证诊断事件会出现在浏览器开发者工具中，同时不会默认向控制台添加噪声。

由索引模型根据 Issue 内容生成。